Figure 30-1 Hemodialysis system. (A) Blood from an artery is pumped into (B), a dialyzer where it flows through

the cellophane tubes, which act as the semipermeable membrane (inset). The dialysate, which has the same

chemical composition as the blood except for urea and waste products, flows in around the tubules. The waste

products in the blood diffuse through the semipermeable membrane into the dialysate. Adapted with permission

from Smeltzer SC, Bare BG. Textbook of Medical-Surgical Nursing. 9th ed. Philadelphia, PA: Lippincott Williams

& Wilkins; 2000.

Solutes from the blood are removed through diffusion and convection. Diffusion is

the process whereby the molecule moves across its concentration gradient by passing

through pores in the dialysis membrane.

7 Once the concentration of a solute reaches

equilibrium on both sides of the membrane, the net movement is zero because the rate

of movement from the blood to dialysate compartment is equal to the rate from the

dialysate to the blood compartment. For most substances, equilibrium is not

achieved, either because the blood and dialysate flow rates are too rapid or the

molecule is too large to easily move through the pores.

Accumulated water is removed by the process of ultrafiltration. A controlled

pressure difference across the semipermeable membrane permits water movement

through the membrane pores, carrying solute into the dialysate, thereby further

enhancing solute removal. Flux is the rate of water transfer across the dialyzer.

Convection is the process that removes toxins and other dissolved solutes during

dialysis through the ultrafiltration of plasma water from the blood compartment. The

removal of solutes by convection during ultrafiltration generally is small relative to

their elimination through diffusion.

DIALYZER CHARACTERISTICS

Dialyzers are characterized by many factors, such as membrane composition, size,

and ability to clear solutes. Their primary component is the dialysis membrane,

which is made of cellulose (e.g., cuprammonium cellulose), substituted cellulose

(e.g., cellulose acetate, cellulose triacetate), cellulosynthetic, or synthetic polymer

(e.g., polysulfone, polyacrylonitrile, and polymethylmethacrylate).

8 Membranes differ

not only by composition but also by surface area, thickness, and configuration within

the dialyzer. The most common configuration is the hollow fiber dialyzer, whereby

the membrane is formed as thousands of hollow fibers that run the length of the

dialyzer. Blood flows through the fibers and the dialysate flows in the space

surrounding the fibers within the dialyzer cartridge. The result is an extremely large

surface area for diffusion, which is functionally increased further by the movement of

blood and dialysate in opposite directions, so that equilibrium is never fully

achieved. Another, less common design is the parallel-plate configuration, whereby

blood and dialysate flow between alternating sheets of the membrane.

Functionally, dialysis filters can be differentiated based on their ability to remove

solutes and water. The flux of water across the dialyzer is correlated with the

clearance of middle molecular weight molecules. Thus, dialyzers are characterized

as low-flux or high-flux based on pore size and ability to remove small versus large

molecules. One method of categorizing and comparing efficiency (flux) of dialyzer

units is to compare the relative in vitro and in vivo clearance rates of marker solutes

of varying molecular size. This information is usually printed on the outside of the

dialyzer or in the package insert (specification

p. 653

p. 654

chart) for the dialyzer. For example, urea (molecular size, 60 Da) is a marker of

small-molecule transport across the dialysis membrane. Urea (found in the blood as

blood urea nitrogen [BUN]) distributes freely throughout body water and is cleared

rapidly by HD, even when using standard low-flux dialyzers. Because the pore size

of most dialyzer membranes is large enough to allow this small molecule to freely

diffuse, the rate-limiting step for the removal of urea is blood flow through the

dialyzer. A larger molecule, vitamin B12

(molecular size, 1,355 Da), has also been

used as a measure of dialysis efficiency. Because vitamin B12

is too large to easily

cross through the pores of conventional dialysis membranes, its dialysis clearance is

less dependent on blood flow than urea. Instead, the overall removal of vitamin B12

depends more on the type of membrane (i.e., thickness and pore size) and the duration

of dialysis. The clearance of β2

-microglobulin, an even larger molecule than vitamin

B12

(molecular size, 11,800 Da), has been used to characterize the flux of a

dialyzer.

4,8 High-flux dialyzers are defined as providing β2

-microglobulin clearances

of at least 20 mL/minute.

4 β2

-microglobulin clearance, however, is not consistently

reported in all dialyzer specification charts. High-flux membranes have larger pores

and are able to clear larger molecules (e.g., middle molecules such as β2

-

microglobulin and leptin) and drugs (e.g., vancomycin or vitamin B12

, with molecular

weights in the range of 1,000–5,000 Da) more effectively than low-flux membranes

with smaller pores. High-flux membranes also have a greater permeability to water,

as reflected in a KUf value (to be defined later) of more than 10 mL/hour/mm Hg.

Similarly, molecular weight of drugs is a predictor of dialysis clearance. At a

molecular weight of less than 500 Da (e.g., aminoglycosides and theophylline),

dialyzability is expected to be high. For these drugs, the actual amount dialyzed will

vary based on protein binding (i.e., amount of unbound drug available to cross the

dialysis membrane), volume of distribution (Vd) (i.e., a large Vd indicates a

relatively small amount of drug will be available in the blood for dialysis), blood

flow rate through the dialyzer, dialysis flow rate, and dialyzer surface area. Drugs

with a molecular weight between 500 and 1,000 Da (e.g., morphine and digoxin) are

less well dialyzed. For digoxin, a greater problem is its large Vd and relatively low

serum concentrations. Even if the drug in the blood is effectively removed, tissuebound drug will quickly redistribute back into the blood as soon as dialysis is

completed, a phenomenon known as rebound. Finally, large molecular weight drugs,

such as vancomycin, are poorly dialyzed by conventional dialyzers, but they may be

removed using high-flux techniques described later in this chapter.

The efficiency of a dialyzer is also a function of its surface area. High-efficiency

membranes generally have a large surface area and are able to clear large quantities

of small molecules, such as urea. High-efficiency dialyzers can also have small or

large pores, resulting in low or high clearance of larger molecular weight solutes.

Membranes also differ in their degree of biocompatibility. When free hydroxyl

radicals on the surface of cellulose membranes come in contact with blood, the

complement pathway is activated and proinflammatory cytokines are produced,

which can lead to hypotension, fever, and platelet activation in patients.

9 Use of these

membranes has declined. The free hydroxyl groups can be substituted with other

chemical structures, such as acetate, to improve biocompatibility. Complement

activation and cytokine release occur to a much lesser extent with substituted

cellulose or cellulosynthetic membranes, and least of all with synthetic membranes

made from plastics.

A typical package insert for a dialyzer will provide information on the clearance

of various molecules (e.g., urea, creatinine, phosphate, and vitamin B12

). Urea

clearance has become a common measure of comparison for membranes; however,

clearance also depends on other factors, such as blood and dialysate flow rates. A

more standard measure for comparison is KoAurea

, the mass transfer area coefficient

for urea. Based on the urea clearance data from the package insert, KoAurea can be

estimated based on blood flow. Using this information, the dialysis prescription can

be individualized to provide a specified dose of dialysis for the patient.

Patients having chronic HD typically are dialyzed for 3 to 4 hours, 3 times a week,

either Monday-Wednesday-Friday or Tuesday-Thursday-Saturday. During the

interdialytic period, fluids ingested and produced through metabolic processes are

retained in the patient. Although patients generally are on fluid-restricted diets,

accumulation of 1 to 5 L of fluid (translating into 1- to 5-kg weight gain) between

sessions is common and must be removed during the dialysis treatment.

BLOOD AND DIALYSATE FLOW

Although small-molecule clearance is highly dependent on blood flow, the

relationship is not strictly linear. Increased blood flow yields a less than

proportional response in urea clearance.

10 This is likely because of an insufficient

time for equilibration to occur between the blood and dialysate compartments as well

as a greater membrane resistance to diffusion from an increased stagnant layer. A

typical blood flow rate for dialysis is 400 to 500 mL/minute, but it is dependent on

the vascular access site and the cardiovascular status of the patient. Some patients

are not able to tolerate this rate, and a lower blood flow rate may be necessary.

Dialysate flow rates generally are 500 mL/minute and can be increased to 800

mL/minute for high-flux dialysis, which will increase urea clearance by

approximately 10%.

11

CASE 30-1

QUESTION 1: R.W., a 55-year-old man with stage 4 CKD as a result of poorly controlled hypertension,

presents to the renal clinic for reassessment of his kidney function. He is 70 inches tall and weighs 70 kg. Since

his last visit 3 months ago, his creatinine clearance (ClCr) has decreased from 22 to 12 mL/minute and the

BUN has increased to 89 mg/dL. The serum potassium (K) is 4.8 mEq/L and HCO3

is 17 mEq/L. He has

selected HD as his form of therapy until a suitable donor kidney is available and is expected to begin dialysis

within the next 1 to 3 months. When he begins dialysis, he will be dialyzed 3 times a week for 4 hours each

treatment, using a Fresenius F-160 dialyzer, with blood and dialysate flows of 400 and 500 mL/minute,

respectively, and bicarbonate-containing dialysate. What characteristics of the Fresenius F-160 dialyzer make it

a good choice for R.W.? What determines the composition of the dialysate?

The Fresenius dialyzer is a high-flux dialyzer as described previously. This

polysulfone membrane is a synthetic membrane with larger pore sizes than

conventional cellulose membranes. The F-160 has a KUf (the ultrafiltration

coefficient [volume of water removed/mm Hg across the membrane per hour of

dialysis]) of 45 mL/mm Hg/hour, indicating a high ultrafiltration capability; an in

vitro KoAurea of 1,064, a measure of dialyzer efficiency for urea removal; urea

clearance of 266 mL/minute at a blood flow of 300 mL/minute; and a surface area of

1.5 m2

.

8 This information can be located in the product literature from the

manufacturer or summary tables from common dialysis references.

8 These data are

used to individualize the dialysis prescription for a patient.

DIALYSATE COMPOSITION

Dialysate composition usually is standardized within certain limits of electrolyte

content, yet allows for individualization as necessary. Water is obtained through the

public water system, which then undergoes treatment by reverse osmosis, followed

by ion exchange with activated charcoal to remove contaminants, such as aluminum,

copper, and chloramines, as well as bacteria and endotoxins.

12 The dialysate solution

does not require sterilization because the dialysis membrane separates the blood and

dialysate compartments. Nevertheless, pyrogen reactions may occur, and a greater

risk may exist with high-flux membranes because of the increased pore size.

p. 654

p. 655

Table 30-1

Electrolyte Composition of Hemodialysis and CAPD Dialysate Solutions

Solute Hemodialysis (mEq/L) CAPD (mEq/L)

Sodium 135–145 132

Potassium 0–4 0

Calcium 2.5–3.5 3.5

Magnesium 0.5–1.0 1.5

Chloride 100–124 102

Bicarbonate 30–38

Lactate 35

pH 7.1–7.3 5.5

CAPD, continuous ambulatory peritoneal dialysis.

The final dialysate solution is prepared in the dialysis machine by proportioning a

dialysate concentrate with the purified water, resulting in a final product, which

typically contains those elements listed in Table 30-1. By adjusting electrolyte

concentration in the dialysate, the efficiency of dialysis for particular chemicals can

be manipulated. For example, if the patient is hyperkalemic, the dialysate contains a

low concentration of potassium for diffusion of potassium from blood into dialysate.

On the other hand, if the patient is normokalemic at the start of dialysis, the potassium

concentration of the dialysate is set at a normal physiologic concentration to

minimize flux of this electrolyte across the membrane. If the concentration of a solute

is higher in the dialysate than in the blood, the net movement will be into the blood,

not out. Metabolic acidosis, which is associated with ESRD because of an inability

to excrete the daily obligatory load of acid, is controlled with the addition of

bicarbonate buffer to the dialysate solution. Before delivery, dialysate is heated to

37°C to maintain body temperature and avoid hemolysis, which can occur with

excessive heating. Dialysate is also deaerated under vacuum to remove dissolved air

from the solution.

Vascular Access

CASE 30-1, QUESTION 2: To achieve a sufficient blood flow for dialysis, R.W. must have a vascular site

for chronic access. What are the options for chronic vascular access in R.W.?

A permanent vascular access site provides easy access to high blood flow, which

cannot be achieved through routine venipuncture of superficial veins. Different types

of vascular access are available: arteriovenous (AV) fistula made by joining an

artery and vein in the arm, AV graft, a soft tube made of polytetrafluoroethylene to

join an artery and vein in the arm, and a double-lumen or tunneled, cuffed catheter

placed in a large vein usually in the neck. AV fistulas and grafts are placed in the

nondominant arm. Ideal vascular access delivers blood flow rates necessary for

chronic HD, has a long period of use, and has a low rate of complications (e.g.,

infection, stenosis, thrombosis, aneurysm, and limb ischemia).

An AV fistula is created surgically by subcutaneous anastomosis of an artery to an

adjacent vein. The AV fistula may not be suitable for patients with poor vasculature,

such as elderly patients or those with diabetes, atherosclerosis, or small vessels. The

K/DOQI guidelines for vascular access advocate placement of a fistula at the

location of the wrist (radial-cephalic), or secondarily the elbow (brachial-cephalic),

as the preferred vascular access sites. Once created, vascular access requires time to

mature before it can be used for HD. The fistula should preferably be created 3 to 4

months before its intended use to allow the vein to mature. The graft can be used soon

after insertion, although 2 weeks will allow for healing at the anastomosis sites and

may prolong patency. AV fistulas fail to mature at a higher rate than grafts; however,

grafts require fourfold higher interventions per year (elective angioplasty,

thrombectomy, or surgical revision) to maintain long-term patency for HD.

13 Central

venous catheters are discouraged for chronic vascular access because of high rates of

infection and occlusion.

During the dialysis procedure, one needle or catheter is placed into the fistula site

to deliver blood to the dialyzer. This is often referred to as the “arterial line” to the

dialyzer. Blood exiting the dialyzer is returned back to the patient’s fistula site

through a second catheter and needle, referred to as the “venous line” from the

dialyzer.

If R.W. has adequate vasculature, a fistula should be created for chronic access.

Vascular access is critical for chronic HD and often has been labeled the Achilles’

heel of dialysis therapy. Complications associated with vascular access are a

significant problem in patients having chronic HD. The most common is thrombosis,

usually the result of venous stenosis.

6

If not treated, thromboses will result in loss of

the access. Access-related complications are a major cause of hospitalization and,

therefore, attention to these problems is important both clinically and economically.

ANTICOAGULATION

CASE 30-1, QUESTION 3: Recommend a reasonable anticoagulation regimen for R.W. with the initiation of

his HD. What are alternatives for patients at high risk for bleeding?

Most patients undergoing HD are anticoagulated with IV heparin during the

dialysis treatment. Anticoagulation is necessary to prevent blood from clotting in the

extracorporeal circuit. Several methods have been used in an attempt to provide

adequate anticoagulation without increasing the risk of bleeding. Approaches include

the administration of heparin in adequate quantities to anticoagulate the patient during

the dialysis procedure by either intermittent bolus injections or an initial bolus

followed by a continuous infusion.

14 Modern HD delivery systems have incorporated

heparin infusion devices that can be programmed to provide the desired infusion rate

during dialysis.

With no evidence of a bleeding disorder, recent surgery, or other risk factors for

heparin anticoagulation, therapy should be initiated with a 2,000-unit bolus of IV

heparin 3 to 5 minutes before initiation of dialysis, followed by an infusion of 1,200

units/hour.

14 The target activated clotting time (ACT) is 40% to 80% above the

average baseline for the dialysis unit (e.g., 200–250 seconds, for normal values of

120–150 seconds). The clinician should monitor for signs of bleeding and measure

the ACT at 1-hour intervals during dialysis. Heparin should be discontinued 1 hour

before the end of dialysis to prevent excessive bleeding after dialysis. Using these

standard doses, the estimated elimination half-life for heparin is approximately 50

minutes, and it should have a linear dose–response relationship within the target

ACT.

14

Patients at increased risk of bleeding include those who have had recent surgery,

retinopathy, gastrointestinal bleeding, and cerebrovascular bleeding. For these

patients, the goal is to prevent clot formation within the dialysis circuit as well as to

minimize the risk of active bleeding. This may be accomplished by using “minimaldose” heparin (tight ACT control), or even heparin-free

p. 655

p. 656

anticoagulation. The minimal-dose heparin approach individualizes therapy to

achieve ACT values 40% above baseline after an initial bolus of 750 units.

14 The

ACT is measured 3 minutes after the bolus dose, which should allow for vascular

distribution of the heparin to be complete. If the goal ACT level is not achieved,

repeat bolus doses of heparin can be administered at a dose that is adjusted based on

the expectation of a linear response. For example, if the first dose of 750 units

reaches 75% of the ACT goal, an additional 250 units would be appropriate for the

second dose. Similarly, the initial heparin maintenance infusion rate of 600 units/hour

can be modified by monitoring the ACT at 30-minute intervals. Adjustments in the

infusion rate should be proportionate to the bolus dose needed to maintain the ACT at

40% above baseline. Samples collected for determination of ACT should be

obtained from the arterial line into the dialyzer, before the infusion of heparin, to

reflect systemic anticoagulation effects.

Heparin-free dialysis is an alternative to heparinization for hemodialysis patients

who are at a moderate-to-high risk of bleeding or who are actively bleeding.

14 This

approach requires priming the hemodialysis circuit and dialyzer with heparin 3,000

units/L in normal saline to coat the extracorporeal surfaces. The heparin-containing

priming fluid is allowed to drain by filling the circuit with either the patient’s blood

or normal saline alone at the outset of dialysis. Next, hemodialysis is set at a high

blood flow rate of 300 to 400 mL/minute, if tolerated. During dialysis, the dialyzer is

flushed with normal saline every 15 to 30 minutes to rinse away microclots that may

have formed. The incidence of clotting with this approach is approximately 5%.

The regional administration of trisodium citrate through the arterial line is an

alternative to systemic anticoagulation. It binds free calcium, which is necessary for

the coagulation process. The calcium citrate complex is removed by the dialysate

and, based on plasma calcium values, calcium chloride is administered on the venous

side to replace the citrate-bound calcium to prevent hypocalcemia or hypercalcemia.

Some of the administered citrate is returned to the patient and is metabolized to

bicarbonate, leading to metabolic alkalosis in some cases. Trisodium citrate may

lead to hypernatremia. Regional citrate anticoagulation is reserved for patients who

are at risk for bleeding and requires additional monitoring to adjust the dual

infusions.

14

In a prospective study of 1,009 consecutive high-flux dialysis procedures

in 59 patients, long-term citrate anticoagulation achieved excellent anticoagulation

(99.6%) with rare (0.2%) adverse effects on ionized calcium levels, electrolytes,

and acid–base balance.